Image enhancement using morphological contrast enhancement for video based image analysis

Image enhancement using morphological operations for video based image is proposed in this paper. Morphological operations used are top-hat and bottom-hat transformation. Top-hat transformation is used to extract the bright areas, and bottom-hat transformation is used to extract dark areas. Contrast enhancement processed by adding bright areas to original image and subtracting by dark areas. Morphological contrast enhancement applied to the preprocessing to generate output image for segmentation process. In this study, it conducted a comparison of the preprocessing based on the quality of clusters produced through the segmentation process. Silhouette coefficient is used to measure the validity of cluster generated in the segmentation process. The silhouette width value of the segmentation results are closed to 1 that expressed as a good cluster. Silhouette value of morphological contrast enhancement is higher than histogram equalization method. Silhouette width value of each preprocessing was 0.7729 and 0.5794, for morphological and CHE, respectively.